Suwanwiwat, Hemmaphan, Blumenstein, Michael, and Pal, Umapada (2015) Short answer question examination using an automatic off-line handwriting recognition system and a novel combined feature. In: Proceedings of the International Joint Conference on Neural Networks. pp. 1-8. From: IJCNN 2015: International Joint Conference on Neural Networks, 12-17 July 2015, Killarney, Ireland.

PDF (Published Version) - Published Version Restricted to Repository staff only

Abstract

Off-line automatic assessment systems can be an aid for teachers in the marking process. There has been no recent work in the development of off-line automatic assessment systems using handwriting recognition, even though such systems will clearly benefit the education sector. The reason is many schools and universities in many parts of the world still use paper-based examination. This research proposes the use of a newly developed feature extraction technique called the Modified Water Reservoir, Loop and Gaussian Grid Feature, as well as other feature extraction techniques. These techniques were investigated employing artificial neural networks and support vector machines as classifiers to develop an automatic assessment system for marking short answer questions. The system has high assessment accuracy (up to 94.75% for hand printed, 96.09% for cursive handwritten, and 95.71% for hand printed and cursive handwritten combined). The proposed system also includes assessment criteria to augment its accuracy.

Item ID:	46486
Item Type:	Conference Item (Research - E1)
ISBN:	978-1-4799-1959-8
Related URLs:	Publisher
Date Deposited:	28 Nov 2016 00:29
FoR Codes:	08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing > 080199 Artificial Intelligence and Image Processing not elsewhere classified @ 80% 08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing > 080106 Image Processing @ 20%
SEO Codes:	89 INFORMATION AND COMMUNICATION SERVICES > 8999 Other Information and Communication Services > 899999 Information and Communication Services not elsewhere classified @ 100%
Downloads:	Total: 1
	More Statistics